[bio_tt]

hi,
bowtie is very popular,but i am a newer.and you say"Includes tool to convert Bowtie output to a Maq .map file ",but i don't know how to write the command line?could you help me? thanks

[gen2prot]

Hello Ben,

I have a question with Bowtie. Does the bowtie-build command create a different index everytime it is run on the same data? Myself and my colleague are creating an index of the Drosophila genome. But he says that the builds are different everytime one runs it. Is this true? If so how is the downstream process (Tophat, Cufflinks) outputs comparable to each other.

Thanks
Abhijit

[Ben Langmead]

I have a question with Bowtie. Does the bowtie-build command create a different index everytime it is run on the same data? Myself and my colleague are creating an index of the Drosophila genome. But he says that the builds are different everytime one runs it. Is this true? If so how is the downstream process (Tophat, Cufflinks) outputs comparable to each other.

Hi gen2prot,

If you run Bowtie twice on the same exact set of fasta files (same sequence names and sequence characters), specified in the same exact order, you should get the same exact index. If you don't, there's something wrong. If you have an example where it doesn't (the smaller the better), please email me.

Thanks,
Ben

[miltron]

Hi, I'm a new member and I'm trying to use Bowtie to align some ChIPseq data. First of all, I'm really grateful that you are putting your efforts into this and really keeping an eye on this board it seems.

I installed Bowtie binaries for Windows on my Windows 7 system and I'm trying to use Cygwin to run it. I have 4GB of RAM installed on this computer. According to the documentation I thought that should have been appropriate memory for my purpose, but I might be understanding.

I'm assembling reads from 2.5 GB file to Mouse mm9 assembly and using the pre-built indexes reccommended in the Bowtie manual.

However, when I run bowtie I get the following error message:

Time loading forward index: 00:00:14
Out of memory allocating the ebwt[] array for the Bowtie index. Please try
again on a computer with more memory.
Time loading mirror index: 00:00:01
Time searching: 00:00:15
Overall time: 00:00:15
Command: C:\users\erna\desktop\bowtie-0.12.5-win32\bowtie-0.12.5\bowtie.exe -t -
m 1 -v 2 -q --best mm9 reads/seq3.fastq seq3align.map


I've tried to increase the memory allocated to Cygwin to no avail. Do you think that finding another Windows system with more memory will help?
Thanks!

[Ben Langmead]

Hi Miltron,

I would recommend trying --offrate 6 and, if that doesn't work, --offrate 7. Those will use a sparser sample of the suffix array than is included in the index by default, which in turn reduces memory footprint. If that doesn't work, I would suggest running on a computer with more memory. The mouse genome is big

Good luck,
Ben

[miltron]

Thanks Ben!
That was a very quick reply! I will try this
Best,
Miltron

[corthay]

Question abou "-n" option.

Hi,

I would like to ask about "-n" option. I would like to map read
following conditions.
1. first N bases must be perfect match.
2. reported position must be 'best' hit.

I used " -n 0 -l N -a --best --strata ". I am wondering if the '--best'
option is only guaranteed "-n" and '-l' option. For example, I worry
that bowtie reports 1 mismatch hit even if there is perfect hit.

Also, I would like to know the option to get random position
if reads hit to multiple position.

Thanks,
Corthay.

[Ben Langmead]

Hi Corthay,

I would like to ask about "-n" option. I would like to map read
following conditions.
1. first N bases must be perfect match.
2. reported position must be 'best' hit.

I used " -n 0 -l N -a --best --strata ". I am wondering if the '--best'
option is only guaranteed "-n" and '-l' option. For example, I worry
that bowtie reports 1 mismatch hit even if there is perfect hit.

--best = best in terms of stratum first and quality second. It should report all 0-penalty alignments before any non-0 penalty alignments. So it should do what you want.

Also, I would like to know the option to get random position if reads hit to multiple position.

That's the default - you'll get that behavior if you omit the -a option.

Thanks,
Ben

[corthay]

Hi Ben,

I appreciate your quick response.
I think I understood it.

Thanks,
Corthay

[didymos]

problem with SOLiD data mapping

Hi,
I am trying to map short reads from SOLiD into indexes generated from stem-loop sequences of miRs to get information about miRNA in my samples.
Indexes I have generated from fasta file (from miRBASE) by bowtie-build -C and it works great. However I have some problem with mapping... I will explain it with example:
mmu mir-378 -> ACTGGACTTGGAGTCAGAAGG
when I have converted it into colorspece (with T at the begining - from promotor) I got:
T312102120102212122020
Mapping this miRNA works great:
bowtie -a -n 0 -C ../mouse/miRNA-stem-loop -c T312102120102212122020

0 + mmT-mir-378 43 CTGGACTTGGAGTCAGAAG qqqqqqqqqqqqqqqqqqq 0


However in my reads from SOLiD I have reads of length 35 so this miRNA is longer and is:
T31210212010221212202033002010303111
but first 23 "numbers" are the same so using this option:
bowtie -a -n 0 -l 16 -C ../mouse/miRNA-stem-loop -c T31210212010221212202033002010303111
I should be able to map this read, but it does not work like this:
# reads processed: 1
# reads with at least one reported alignment: 0 (0.00%)
# reads that failed to align: 1 (100.00%)
No alignments

and I have no idea why...
Thank you for any sugestions!
Best

tomek

[afadda]

hi. i get segmentation fault only when I run bwotie view on a certain data. it always stops at the same genome coordinates, whether it's in one big file with all chromosomes in, or just a file of that particular chromosome. can't tell what the problem is. help!!

[Ben Langmead]

bowtie view? I'm confused - can you send the exact command you're using?

Thanks,
Ben

hi. i get segmentation fault only when I run bwotie view on a certain data. it always stops at the same genome coordinates, whether it's in one big file with all chromosomes in, or just a file of that particular chromosome. can't tell what the problem is. help!!

[Ben Langmead]

However in my reads from SOLiD I have reads of length 35 so this miRNA is longer and is:
T31210212010221212202033002010303111
but first 23 "numbers" are the same so using this option:
bowtie -a -n 0 -l 16 -C ../mouse/miRNA-stem-loop -c T31210212010221212202033002010303111
I should be able to map this read, but it does not work like this:
# reads processed: 1
# reads with at least one reported alignment: 0 (0.00%)
# reads that failed to align: 1 (100.00%)
No alignments

and I have no idea why...
Thank you for any sugestions!

I think the problem is that you're assuming -n 0 -l 16 is going to allow any alignment with no mismatches in the first 16 colors, but the -e limit also applies. -e also needs to be set high enough so that the sum of the quality values of all the mismatched positions in your example are still <= the -e limit. If no quality values are supplied, qualities all = 30.

Hope that helps. You may also want to consider using trimming (e.g. -3).

Thanks,
Ben

[afadda]

bowtie view? I'm confused - can you send the exact command you're using?

Thanks,
Ben

sorry!! I actually meant samtools view. perhaps this is not the right string for my problem.
thanks

[issaac]

calling variants with samtools after bowtie -- very long lines in output

Hi,

I am facing the following situation. It looks like there could be an error in my approach, but I am trying to figure out. I would much appreciate your inputs/suggestions to resolve my problem.

1. (PROBLEM) I am trying to align PhiX-174 reads, obtained from GA IIx, on to the reference (from NCBI) using the bowtie program and then subsequently look for variants using samtools.

I am getting very long lines in the final SNP calling output file and the result doesn't look meaningful.

2. Illumina-analysis-software-generated fastq file, which I am using contains some 18 million read sequences.

3. If needed, I can provide a truncated version of the output, as the output is 366M in size.

4. To check the Bowtie installation on my computer, I ran the example provided in the Bowtie: Tutorial section, especially, Finding variations with samtools -- the run was successful.

I performed the following steps:

STEP #1: Generate Indexes for the PhiX-174 genome
bowtie-build phix174.fasta phix174

STEP #2: Perform alignment with bowtie and generate sam file
bowtie -S phix174 phix174.fastq phix174.sam

STEP #3: Convert sam to bam file
samtools view -bS -o phix174.bam phix174.sam

STEP #4: Sort the bam file in preparation for SNP calling
samtools sort phix174.bam phix174.sorted

STEP #5: Identify SNPs...
samtools pileup -cv -f phix174.fasta phix174.sorted.bam > phix174.variants

Thanks,
Issaac